The R of the RC puts an additional zero in the compensation network. This zero should be matched to a parasitic pole coming from the folded cascode. The M21/M22 have a transconductance (gm in DCOP result file) which make with the CGS and CSB and other caps there a parasitic pole. Because this pole is bias and tech dependend most often the R is made of an NMOS and matched to the gm above.
By the way the common voltage of the differential folded cascode is undefined. There two unrelated pairs of curent sources act against!
R=1/gm51 to cancell the Influence of the zero point and C depends on the load capacitor and the ratio of the first stage amplifier gain and the second stage amplifer.usually we take the factor of 2~3 .
i'm sorry to tell you that this paper is crap. i read this paper before. i useed all of the parameters and sizing the author used but i didnt get what he has on this paper.
btw, are you designing a telescopic or a folded cascode?
i'm sorry to tell you that this paper is crap. i read this paper before. i useed all of the parameters and sizing the author used but i didnt get what he has on this paper.
btw, are you designing a telescopic or a folded cascode?
Sometimes...R value might not need it. (potentially with more power consumption, but negligibly small)
C should be determined by understanding between 1st and 2nd stage power consideration.
For example, if C is large, it splits poles farther apart (thus, better in stability), however, 1st stage needs lots of power to have reasonable bandwidth.
R value is easy to determined....R is used to push a right half plane zero to left half plane zero or to the infinity.
yes i am designing telescopic op amp, could you help me?
do u mean this paper
"a 1.0v 10b 100 MS/s pipeline ADC in 90 nm cmos"
**************************************************
Hi,
711_proj.pdf isn't a good reference.
you need help with sizing the fets? it would be better if you post out the power supply and Vov you assign for each of the transistor.